1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
|
/* Tests the bug that caused the phi nodes not be replicated (launchpad #927573).
Copyright (c) 2012 Pekka Jääskeläinen / Tampere University of Technology
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/
#include "pocl_opencl.h"
// Enable OpenCL C++ exceptions
#define CL_HPP_ENABLE_EXCEPTIONS
#define CL_HPP_MINIMUM_OPENCL_VERSION 120
#define CL_HPP_TARGET_OPENCL_VERSION 120
#include <CL/opencl.hpp>
#include <cstdio>
#include <cstdlib>
#include <iostream>
#define WINDOW_SIZE 16
#define WORK_ITEMS 16
#define BUFFER_SIZE (WORK_ITEMS + WINDOW_SIZE)
float A[BUFFER_SIZE];
int R[WORK_ITEMS];
static char
kernelSourceCode[] =
"kernel \n"
"void test_kernel(__global float *input, \n"
" __global int *result) {\n"
" int gid = get_global_id(0);\n"
" float global_sum = 0.0f;\n"
" int i;\n"
"\n"
" for (i=0; i < 16; ++i) {\n"
" float value = input[gid+i];\n"
" global_sum += value;\n"
" barrier(CLK_LOCAL_MEM_FENCE);\n"
" }\n"
" result[gid] = global_sum;\n"
"}\n";
int
main(void)
{
for (int i = 0; i < BUFFER_SIZE; i++) {
A[i] = (float)i;
}
for (int i = 0; i < WORK_ITEMS; i++) {
R[i] = i;
}
std::vector<cl::Platform> platformList;
bool ok = false;
try {
// Pick platform
cl::Platform::get(&platformList);
// Pick first platform
cl_context_properties cprops[] = {
CL_CONTEXT_PLATFORM, (cl_context_properties)(platformList[0])(), 0};
cl::Context context(CL_DEVICE_TYPE_ALL, cprops);
// Query the set of devices attched to the context
std::vector<cl::Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
// Create and program from source
cl::Program::Sources sources({kernelSourceCode});
cl::Program program(context, sources);
// Build program
program.build(devices);
// Create buffer for A and copy host contents
cl::Buffer aBuffer = cl::Buffer(
context,
CL_MEM_READ_ONLY | CL_MEM_COPY_HOST_PTR,
BUFFER_SIZE * sizeof(float),
(void *) &A[0]);
// Create buffer for that uses the host ptr C
cl::Buffer cBuffer = cl::Buffer(
context,
CL_MEM_WRITE_ONLY | CL_MEM_USE_HOST_PTR,
WORK_ITEMS * sizeof(int),
(void *) &R[0]);
// Create kernel object
cl::Kernel kernel(program, "test_kernel");
// Set kernel args
kernel.setArg(0, aBuffer);
kernel.setArg(1, cBuffer);
// Create command queue
cl::CommandQueue queue(context, devices[0], 0);
// Do the work
queue.enqueueNDRangeKernel(
kernel,
cl::NullRange,
cl::NDRange(WORK_ITEMS),
cl::NullRange);
// Map cBuffer to host pointer. This enforces a sync with
// the host backing space, remember we choose GPU device.
int * output = (int *) queue.enqueueMapBuffer(
cBuffer,
CL_TRUE, // block
CL_MAP_READ,
0,
WORK_ITEMS * sizeof(int));
ok = true;
for (int i = 0; i < WORK_ITEMS; i++) {
float global_sum = 0.0f;
for (int j=0; j < 16; ++j) {
float value = A[i + j];
global_sum += value;
}
if ((int)global_sum != R[i]) {
std::cout << "F";
ok = false;
}
}
// Finally release our hold on accessing the memory
queue.enqueueUnmapMemObject(
cBuffer,
(void *) output);
queue.finish();
}
catch (cl::Error &err) {
std::cerr << "ERROR: " << err.what() << "(" << err.err() << ")"
<< std::endl;
return EXIT_FAILURE;
}
platformList[0].unloadCompiler();
if (ok) {
std::cout << "OK" << std::endl;
return EXIT_SUCCESS;
} else {
std::cout << "FAIL" << std::endl;
return EXIT_FAILURE;
}
}
|
